This Title All WIREs
How to cite this WIREs title:
WIREs Nanomed Nanobiotechnol
Impact Factor: 9.182

Dark‐field hyperspectral imaging for label free detection of nano‐bio‐materials

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract Nanomaterials are playing an increasingly important role in cancer diagnosis and treatment. Nanoparticle (NP)‐based technologies have been utilized for targeted drug delivery during chemotherapies, photodynamic therapy, and immunotherapy. Another active area of research is the toxicity studies of these nanomaterials to understand the cellular uptake and transport of these materials in cells, tissues, and environment. Traditional techniques such as transmission electron microscopy, and mass spectrometry to analyze NP‐based cellular transport or toxicity effect are expensive, require extensive sample preparation, and are low‐throughput. Dark‐field hyperspectral imaging (DF‐HSI), an integration of spectroscopy and microscopy/imaging, provides the ability to investigate cellular transport of these NPs and to quantify the distribution of them within bio‐materials. DF‐HSI also offers versatility in non‐invasively monitoring microorganisms, single cell, and proteins. DF‐HSI is a low‐cost, label‐free technique that is minimally invasive and is a viable choice for obtaining high‐throughput quantitative molecular analyses. Multimodal imaging modalities such as Fourier transform infrared and Raman spectroscopy are also being integrated with HSI systems to enable chemical imaging of the samples. HSI technology is being applied in surgeries to obtain molecular information about the tissues in real‐time. This article provides brief overview of fundamental principles of DF‐HSI and its application for nanomaterials, protein‐detection, single‐cell analysis, microbiology, surgical procedures along with technical challenges and future integrative approach with other imaging and measurement modalities. This article is categorized under: Diagnostic Tools > in vitro Nanoparticle‐Based Sensing Diagnostic Tools > in vivo Nanodiagnostics and Imaging Implantable Materials and Surgical Technologies > Nanoscale Tools and Techniques in Surgery
(a) A labeled schematic of the condenser optics in cytoviva hyperspectral imaging system. Reproduced and redrawn with permission from More and Vince (2014) and Oh et al. (2013). (b) Hyperspectral imaging system spectrograph light path and optics (Lu & Fei, 2014)
[ Normal View | Magnified View ]
(a–f) Segmentation of bacterial colonies touching each other (Arrigoni et al., 2017). (a) True color picture of the original dataset; (b) extraction of forefront; (c, d) spatial transform and spectral transform respectively; (e, f) segmentation results respectively for (c, d). For (g–i) total viable counts of the microorganisms can be estimated using hyperspectral imaging on the various meat and meat products. Researchers (Feng & Sun, 2013) were able to produce prediction maps for the chicken sample by applying partial least squares regression statistical models based on multiple spectral parameters. Black color is used for background false coloring. Color bar on the right indicates varying bacterial loads in log10 colony forming units per gram
[ Normal View | Magnified View ]
(a) Absorption spectra of few important biological constituents across UV–Vis–NIR wavelengths observed in a typical hyperspectral imaging (HSI) studies of biological tissue (Jacques, 2013). (b) Delineating margin of tumor for head and neck cancer using HSI. HSI of cancer tissue (top left), after histological processing of the tissue, tumor margins were highlighted on the pathological image (bottom right), which was utilized to confirm the tumor classification results (top‐right). Bottom left shows the average spectra captured for different types of tissue namely normal tissue and tissue with tumor (Fei et al., 2017). (c) in vivo multispectral imaging/optical microscopy setup to investigate small animals (Panasyuk et al., 2007). (d) Significant information is extracted from identification of hematoma in optical image and in HSI exhibiting multiple entities due to pseudo coloring of pixels with different spectral dataset (Panasyuk et al., 2007). (e) Normalized cancer index and wavelength filtering are evaluated to detect tumor (Akbari et al., 2011). (f) Polarimetric image analysis of blood vessels (Cha et al., 2018). (g) Thematic maps of the hyperspectral imaging where the red color means tumor tissue and green represents normal tissue (Halicek, Fabelo, Ortega, Callico, & Fei, 2019)
[ Normal View | Magnified View ]
(a) Dark‐field hyperspectral imaging (HSI) using plasmonic nanoparticles and quantum dots to study modification of DNA in cells by quantifying 5‐carboxylcytosine. Upper left: unmapped cell image; upper right: mapped spectral image in the threshold 520–620 nm range; lower left: cell imaged at wavelength >635 nm; and lower right: pseudo colored mapped parts of the single cell with Au nanoparticles (shown in figure via green dots) (X. Wang et al., 2015). (b) Upper left: three different conjugated carbon nanoparticles (CNP) mapped via HSI; upper right: molecular bond diagram of prodrug‐CNP; lower left: image of a MCF‐7 breast cancer cell displaying spectrally mapped information of the drug, nanoparticle, and the cell; lower right: localization of drug‐CNP (red pixel, white arrow) in breast cancer with for 4 hr as studied by Misra, Ostadhossein, Daza, Johnson, and Pan (2016). (c) Label‐free HSI of human erythrocytes studied by Conti et al. (2016): upper left: dark‐field image of the human RBC sample; upper right: user‐defined spectral library consisting different endmember spectra with arbitrary color coding; lower left: magnified image of single‐erythrocyte; lower middle: colored pixels showing the matched spectra from the collected spectral library; lower right: image showing the mapping of five major components of red blood cells namely phospholipid, cholesterol, hemoglobin, protoporphyrin, and spectrin. (d) HSI of SHSY5Y cells with iron compounds reprinted from Oh et al. (2013); upper left: SHSY5Y cells continuously incubated with ferric ammonium nitrate for 1 hr; upper right: intensity count plotted against VNIR wavelengths for the marked regions in (upper left). Significant spectral peaks are seen between 450 and 650 nm. Lower: 17‐times magnified HSI images containing data from glass‐substrate, bulk‐iron, nuclei, and cytoplasm
[ Normal View | Magnified View ]
(a) Schematic illustration of quantitative volumetric Raman imaging (qVRI) process to characterize three‐dimension (3D) morphology of THP‐1 cells and THP‐1 differentiated macrophages (Mϕ). Typical confocal microscopy offers capability to obtain images in all 3D. Each 3D imaging plane is represented by the hyperspectral data cube. Therein, each associated 3D data sets are characterized with x × y spatial dimensions and Raman spectrum (w) as spectral dimension. (b) Left: important subcellular structures of cells and macrophages were characterized using label free qVRI method. and right: related endmember Raman spectra shown from top to bottom representing subcellular components like cytoplasm (blue), nucleus (red), triacylglycerols (green), phospholipids (orange), cholesterol (magenta) (Kallepitis et al., 2017). (c) Top: schematic diagram of instrument showing the principles of hyperspectral infrared nanoimaging. It uses a mid‐infrared broadband laser source for illuminating the tip. The probe tip is scanned across the surface of the sample, while at every pixel position the backscattered light from the tip is detected using a Michelson interferometer set up which functions as a Fourier transform spectrometer to generate the IR spectrum. Bottom left: hyperspectral Fourier transform infrared (FTIR) data cubes obtained with a spectral resolution of about 35 cm−1. Bottom right: two‐dimensional array of interferograms to obtain FTIR results. (Amenabar et al., 2017). (d) Applying denoising algorithms to obtain spatial profiles including proteins, nucleic acids and lipids from FTIR image of pancreatic tissue (Wrobel et al., 2019)
[ Normal View | Magnified View ]
(a) Top: schematic diagram for the label‐free optical detection of antibodies/proteins secreted from cell through antibody‐gold nanoparticle interaction using localized surface plasmon resonance (LSPR) imaging. (b) Real‐time monitoring of the nanoplasmonic binding kinetics of a c‐myc peptide functionalized gold nanoparticle arrays to 200 nM of anti‐c‐myc antibodies. (Raphael, Christodoulides, Delehanty, Long, & Byers, 2013). Bottom: (c) scattering characteristics of plasmonic nanostructures before and after sequence specific recognition of DNA binding. Monomers without plasmon coupling (blue line) and dimers with plasmon resonance coupling (purple line) distinguished from each other from their LSPR resonance which is red‐shifted as a result of plasmon resonance coupling in dimers. Top: dark‐field imaging (left panels) and finite‐difference time‐domain simulations (right panels) showing the field intensities of the single nanoparticle or monomer (i) and plasmonic coupled nanoparticle dimer (ii) showing the field intensities. Scale bars, 1 μm (dark‐field microscopy images) and 20 nm (FDTD results) (K. Lee et al., 2014). (d) Top: schematic illustration of LSPR scattering for single miRNA‐21 detection. Plasmonic nanobiosensor is composed of pyramidal tsDNA17‐modified [email protected] nanocubes (NCs) attached on the surface of an ITO coated glass slide, can detect miRNA‐21 with high specificity. Bottom: dark‐field imaging results of [email protected] NC‐tsDNA17 (b‐I) before interacting with miRNA‐21 and [email protected] NC‐tsDNA17‐miR‐21 (b‐II) corresponds to after binding with miRNA‐21 (Y. Zhang et al., 2018)
[ Normal View | Magnified View ]
(a) Left: schematic representation of the time‐dependent visualization of the galvanic exchange (GE) reaction using a dark‐field light scattering microscopy (Jun Zhou et al., 2018). Right: DFM images and spectra of spherical AgNPs during GE reaction (scale bar = 2 μm). (b) Schematic representation of the interaction between gold nanoparticles and CH3NH3PbI3 analysis (bottom) and the corresponding images acquired using DF‐HSI (Xu et al., 2016). (c) Top: two‐directional electron transfer at a Scotty interface comprising of a single CdS nanoparticle and planar gold substrate showing either deposition (Scheme I) or dissolution of sulfur (Scheme II). Bottom: image comparing the deposition and dissolution of sulfur on a single CdS nanoparticle obtained using surface plasmon resonance microscopy (Z. Li, Fang, et al., 2017). (d) Bright‐field, EDFM, and hyperspectral imaging (HSI) of histological porcine skin tissues injected with ceria and alumina nanoparticles represented as engineered nanomaterials. The first column shows a bright‐field image of an H&E stained sample, second column shows EDFM image and third column depicts HSI image where high contrast regions corresponds to ceria and alumina nanoparticles. Column 4 shows the HSI mapping with respect to reference spectral library where blue regions represents ceria and magenta corresponds to alumina nanoparticles (Pena et al., 2016)
[ Normal View | Magnified View ]
(a) Left: schematic showing the mechanism of scattering recovery based plasmonic resonance energy transfer (SR‐PRET); middle: the corresponding dark‐field image of Au NP before (I) and after (II) the addition of F ions, showing the increased scattering intensity of NP after eliminating PRET; right: scattering spectra for the same particle shown in the middle (Shi, Jing, Gu, & Long, 2015). (b) Left: schematic illustration of LSPR phenomenon. (c) Schematic showing the click‐chemistry reactions to form a five‐membered ring structure in the presence of Cu+; right: the click reaction and the corresponding scattering spectra of the NP obtained using DF‐HSI (Shi et al., 2013). (d) (i) Schematic showing the [email protected] core–shell structure (Chen et al., 2015); (ii) schematic of the parameters controlling the plasmon resonance scattering properties of [email protected], where ε1 = core dielectric constant, ε2 = shell dielectric constant and εm = dielectric constant of the surrounding medium; (iii) scattering spectra of a single [email protected] etched using O2˙ for 0, 30, and 60 min. The insets show the optical scattering images of the NP at corresponding time point (scale bar = 500 nm); (iv) TEM images showing the unetched, partially etched, and completely etched NPs (scale bar = 20 nm)
[ Normal View | Magnified View ]
(a) Specimen scan via pinhole enabled point scanning technique, (b) slit‐scanning technique integrated in hyperspectral imaging to perform line scan, (c) monochromator or tunable filter enabled wavelength scan technique, and (d) comprehensive snapshot imaging using array of prisms (Y. W. Wang, Reder, Kang, Glaser, & Liu, 2017)
[ Normal View | Magnified View ]

Browse by Topic

Implantable Materials and Surgical Technologies > Nanoscale Tools and Techniques in Surgery
Diagnostic Tools > In Vivo Nanodiagnostics and Imaging
Diagnostic Tools > In Vitro Nanoparticle-Based Sensing

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts